Download OP-AMP Basics Simple OP-AMP circuits

OP-AMP Basics
Operational amplifiers are convenient building blocks that can be used to build amplifiers, filters, and
even an analog computer. Op-amps are integrated circuits composed of many transistors & resistors such
that the resulting circuit follows a certain set of rules. The most common type of op-amp is the voltage
feedback type and that's what we'll use.
The schematic representation of an op-amp is shown to the left. There are two input
pins (non-inverting and inverting), an output pin, and two power pins. The ideal opamp has infinite gain. It amplifies the voltage difference between the two inputs and
that voltage appears at the output. Without feedback this op-amp would act like a
comparator (i.e. when the non-inverting input is at a higher voltage than the inverting
input the output will be high, when the inputs are reversed the output will be low).
Two rules will let you figure out what most simple op-amp circuits do:
1. No current flows into the input pins (i.e. infinite input impedance)
2. The output voltage will adjust to try and bring the input pins to the same voltage (this rule is
valid for the circuits shown below)
Simple OP-AMP circuits
Voltage Follower:
Vin
Vout
RL
GND
No current flows into the input, Rin = ∞
The output is fed back to the inverting input. Since the
output adjusts to make the inputs the same voltage
Vout = Vin (i.e. a voltage follower, gain = 1).
This circuit is used to buffer a high impedance source
(note: the op-amp has low output impedance 10-100Ω).
Application hint: The input impedance on some CMOS amplifiers is so high that without any input the
non-inverting input can float around to different voltages (i.e. the input pin picks up signals like an
antenna). In cases where the input could be disconnected (such as when coming from an external sensor)
it's a good idea to tie the input to ground via a high value resistor (1M-10M Ω). This keeps the input at
ground potential if the wire to the sensor becomes disconnected and still has high input impedance.
Non-inverting Amplifier:
Vin
Vout
R2
No current flows into the input, Rin = ∞
The output adjusts to bring Vin- to the same voltage as
Vin+. Therefore Vin- = Vin and since no current flows into
Vin- the same current must flow through R1 & R2. Vout is
therefore VR1 + VR2 = Vin- + IR2 = Vin- + (Vin/R1)R2.
Vout = Vin (1+ R2/R1).
Application hint: When dealing with larger signals keep
in mind that the output can't exceed the power supply
R1
voltage (i.e. if the op-amp is powered from +/- 15V and
you have a one volt input and a gain of 20 you won't get
20V on the output. The output will most likely stop a few
volts below the supply rail (around 13V). There are special
GND
op-amps designed to handle inputs and outputs that go all
the way to the power supply rail but these op-amps usually only operate at lower voltages (i.e. 0-5V).
.
R2
R1
Vin
R1||R2
GND
Inverting Amplifier:
Because no current flows into the input pins there can't
be a voltage drop across R1||R2. Vin+ is therefore at 0V
(this is called a virtual ground). The output will adjust
such that Vin- is at zero volts. This makes Rin = R1
Vout
(not ∞). The current through R1 & R2 have to be the
same since no current goes into the input pins.
Therefore I = Vin/R1. Vout = Vin+ - IR2 =
0 - (Vin/R1)R2. Therefore Vout = -Vin(R2/R1). Note:
The negative sine is because the current flows from the
input to the output where as in the earlier examples the
current flows from the output to the input.
Application hint: Why not connect Vin+ directly to
ground? Actually, many people do and the circuit
.
works fine. The reason to have R1||R2 is because real
op-amps aren't perfect and draw a small bias current into both inputs. By adding R1||R2 the voltage drop
at the + input is offset by the same amount as the voltage drop at the - input. This is called input offset
voltage.
R4
R1
V1
R2
V2
R3
V3
R1||R2||R3||R4
Summing amplifier:
Since Vin- is a vertual ground adding V2 and R2
(and V3 & R3) doesn't change the current flowing
through R1 from V1. Each input contributes to
the output using the following equation:
Vout Vout = -V1(R4/R1) - V2(R4/R2) - V3(R4/R3).
The input impedance for the V1 input is still R1,
similarly V2's input impedance is R2 and V3's is
R3. Most of the time the parallel combination of
R1-R4 isn't used and Vin+ is grounded.
Application hint: Some op-amps have null pins
that allow you to add a potentiometer and null
GND
(remove) the input offset error (Ex: The LM741:
http://media.digikey.com/pdf/Data%20Sheets/Fairchild%20PDFs/LM741.pdf). Some newer op-amps (Ex:
LTC1151: http://cds.linear.com/docs/en/datasheet/1151fa.pdf) are chopper stabilized (i.e. they measure
the offset and null it automatically many times a second). Chopper stabilized op-amps work best with
slow moving inputs (temperature for example, something well below the rate at which the offset is
measured and nulled).
.
R2
R1
V1
V2
Difference Amplifier:
You can work out the gain as before using the two
rules (no current flows into the inputs, and the
output will adjust to bring Vin- to Vin+). The
result is Vout = 2(V2-V1)*(R2/R1). Also,
Vout Rin(-) = R1, Rin(+) = R1 + R2.
Application hint: Use precision resistors. If the
two R2 resistors differ by 1% you're difference
will be off by 1%. Same goes for the R1 resistors.
R2
Assume V1 = V2. If the resistors aren't matched
the output won't be zero. Actually, it's better to
buy a difference amplifier or an instrumentation
GND
amplifier instead of building the difference amp
from an op-amp and discreate resistors. By using a difference amp with the resistors on the IC they can
be laser trimmed to better than 0.01% (Ex: LT1190: http://cds.linear.com/docs/en/datasheet/1190fa.pdf, or
LT1168: http://cds.linear.com/docs/en/datasheet/1168fa.pdf).
R1